Please wait a minute...
工程设计学报
工程设计学报  2014, Vol. 21 Issue (5): 405-411    DOI: 10.3785/j.issn. 1006-754X.2014.05.001
保质设计     
模糊支撑半径变量贝叶斯网络多态系统故障概率分析
张路路1，张瑞军1，2，王晓伟1，2，王囡囡1
1.山东建筑大学 机电工程学院，山东 济南 250101；
2.山东省高校机械工程创新技术重点实验室（山东建筑大学），山东 济南 250101
Failure probability analysis of multi-state system based on Bayesian network with fuzzy support radius variable
ZHANG Lu-lu1， ZHANG Rui-jun1，2， WANG Xiao-wei1，2， WANG Nan-nan1
1. School of Mechanical and Electronic Engineering， Shandong Jianzhu University， Jinan 250101, China；
2. Key Laboratory of Mechanical Engineering & Innovation Technology in Universities of Shandong（Shandong Jianzhu University）， Jinan 250101, China
 全文: PDF(1924 KB)   HTML
摘要： 针对现有隶属函数描述系统及部件故障状态的不足，构建了一种含模糊支撑半径变量的隶属函数来描述部件故障状态，提出了一种基于模糊支撑半径变量的贝叶斯网络多态系统故障概率计算方法.在隶属函数的构造中，用变量代替精确值描述隶属函数模糊支撑半径，建立了含模糊支撑半径变量的隶属函数；并将其引入贝叶斯网络，利用桶消元法对多态系统叶节点故障概率进行分析，得到叶节点故障概率变化曲线.将所提出的方法与文献中基于模糊支撑半径为定值的隶属函数贝叶斯网络方法相比较，验证了方法的有效性.最后，将该方法应用到数控机床主轴系统故障分析实例中，对叶节点故障概率进行计算.结果表明，该方法能够有效解决多态系统故障状态隶属函数选择中的主观性问题.
关键词： 模糊支撑半径变量;  隶属函数;  贝叶斯网络;  多态系统;  故障概率;  主轴系统    
Abstract: In order to overcome the shortage of describing the fault states of system and components with existing membership functions, a membership function with fuzzy support radius variable is constructed to describe the fault states and a failure probability calculation method of multistate system based on Bayesian network with variable fuzzy support radius is proposed. The variable was introduced to describe fuzzy support radius of membership function and used to construct the membership function. Introducing the established membership function with variable into Bayesian network, the leaf node failure probability curve was obtained by analyzing the failure probability of leaf nodes on the multistate system using bucket elimination method. The proposed method was compared with Bayesian network method based on membership function with constant fuzzy support radius to verify the validity and generality of the method. Finally, the method was applied to analyze the CNC machine tool spindle system, and calculated the leaf node failure probability. The results show that the method can solve the subjectivity of fault states membership function selection in multistate system effectively.
Key words: fuzzy support radius variable    membership function    Bayesian network    multi-state system    failure probability    spindle system
收稿日期: 2014-06-05 出版日期: 2014-10-28
基金资助:

国家高技术研究发展计划（863计划）资助项目（2014AA041503）；国家自然科学基金资助项目（51175312）
作者简介: 张路路（1988—），男，山东德州人，硕士生，从事机械系统可靠性评估及其故障诊断等研究，E-mail: shuyin123@163.com. 张瑞军（1965—），男，教授，硕士生导师，E-mail: zhangruijun@sdjzu.edu.cn.
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

张路路，张瑞军，王晓伟，王囡囡. 模糊支撑半径变量贝叶斯网络多态系统故障概率分析[J]. 工程设计学报, 2014, 21(5): 405-411.

ZHANG Lu-lu， ZHANG Rui-jun， WANG Xiao-wei， WANG Nan-nan. Failure probability analysis of multi-state system based on Bayesian network with fuzzy support radius variable. Chinese Journal of Engineering Design, 2014, 21(5): 405-411.

链接本文:

https://www.zjujournals.com/gcsjxb/CN/10.3785/j.issn. 1006-754X.2014.05.001    或    https://www.zjujournals.com/gcsjxb/CN/Y2014/V21/I5/405

[1] 邓小雷, 庞世杰, 李瑞琦, 周宜博, 王建臣, 傅建中. 基于昆虫翅脉仿生流道的数控机床主轴系统冷却结构热设计[J]. 工程设计学报, 2018, 25(5): 583-589.
[2] 黄强, 丁志华, 吴崇俊, 刘鑫. 油压扰动的非均匀分布隶属函数模糊控制方法[J]. 工程设计学报, 2007, 14(6): 478-481.